• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Rod Coleman filament supply circuit boards

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2008
Paid Member
Hi,
I started planning a board for a filament supply using Rod Coleman's design when I remembered seeing a post of his mentioning doing a run. 'Trouble is, I can't find it today. Does anybody know if it ever happened, or is my memory just a product of wishful thinking?

Also: Have any others built this supply circuit? Any comments or tips from your own experience?


Thanks very much
 
It's true I was planning on a PCB for the DHT heater. As usual with these things, there are insufficient hours in a day, and it didn't get beyond a notion.

Other more industrious friends on the forum have already created & proved a design, and we are looking at possible ways to make it available. (thanks to Pete)

One possibility is to make the design files available (Eagle based) so that DIYers can forward them to Olimex

http://www.olimex.com/pcb/index.html

who do a good job, including panelisation for a nice price. Panelisation is ideal for this design, since their DSS is only 30 Euros, and you get a good number of DHT heat boards out of that (100x160mm).

It hardly seems worth organising a group buy, when these piece prices are so low.
 
I was waiting for an OK from Rod on this. Piotr and myself did a shootout of filament design possibilities and have made a small run of experimental boards with Rod's consent - they worked just fine and sounded good. Rod's is the best design I've heard out of those I've tried (excludes choke input).

The board we commissioned was in fact quite tiny - half a credit card size. It should be possible, depending from an OK from Piotr who is on holiday, to scare up some more boards from the same source, which Is in Bulgaria if my memory is right. So I would have thought we could get some boards to DIY-ers one way or another.

Andy
 
Sorry Andy! didn't appreciate you were waiting for me.

By all means go ahead and circulate the design through the forum. For DIY noncommercial use of course!

Perhaps if the PCB were a little bigger, we could include some high-power heatsink-mounted resistors for BIG DHT filament heat?

Rod
 
Member
Joined 2008
Paid Member
Thanks very much to both of you !

From Andy - ". . . . . the best design I've heard out of those I've tried (excludes choke input)."

Andy, Did you mean by that that you liked choke input or that you hadn't tried it in the shootout?
I've tried a few LCLC supplies that worked well and what really caught my attention over Rod's design was that I was in the process of wiring up a (described in series) L(cmc) >C> L> Fil> L > Neg. term of C > back out through other leg of CMC. This has a 431 controlled power transistor as a shunt reg. across the fil. It's not finished yet as other stuff is demanding my time but I hope to have it running in the next couple of weeks. (Trying batteries now)

. . . .So I would have thought we could get some boards to DIY-ers one way or another.

I'd certainly be into that if you do!


From Rod - "Perhaps if the PCB were a little bigger, we could include some high-power heatsink-mounted resistors for BIG DHT filament heat?"


I'd very much like to try your design on both low and high power filaments . (When I posted I had been doodling a block layout with transistors and power resistors on the long edge of a narrow strip so that they could be bolted through heat sink pads to the metal side of a chassis and caps in a row down the other side. . . .'don't know yet whether the traces will work out well that way or not.)

Anyway, thanks very much !!
 
Rod Coleman said:
Sorry Andy! didn't appreciate you were waiting for me.

By all means go ahead and circulate the design through the forum. For DIY noncommercial use of course!

Perhaps if the PCB were a little bigger, we could include some high-power heatsink-mounted resistors for BIG DHT filament heat?

Rod

Yes, please publish the design.

TIA,

-- josé k.
 
To answer a few of these questions. I built and listened to a few different filament supplies. I didn't build a choke input one, and haven't heard that as a comparison - I'm sure it's excellent but I need something smaller. As a result of listening tests I and Piotr would rank them as follows:

Rod's design
Current source using LM1084
Voltage regs using LM1084 plus CMC at the end
Voltage regs using LM1084 (no CMC)
Current source using LM1084 plus CMC at the end (didn't sound good)

I built all these on strips of tag board to test out. Piotr and I then discussed doing a prototype PCB with Rod, who was agreeable and most helpful (excellent guy).

Piotr then designed the layout and we commissioned the PCBs from an E.European source that Piotr uses. We used the PCBs successfully in a few projects and they seemed reliable if used within spec. At this stage we've run out of boards, but we could get some more done. They are quite tiny - half a credit card size.

Piotr is away currently so he should be included in all this.

The design of the filament supply is in this forum in this thread

http://www.diyaudio.com/forums/showthread.php?threadid=38248&highlight=

Andy
 
Hi Guys, I've passed the Eagle files to Rod. There are a couple of versions - one with cap + diodes onboard and a smaller one without (so that I could do LC filtering). We've only had the smaller ones made so far. The small board is about 40x40mm. I think we got 48 boards for about 130 Euros (DSQ board).

Olimex will take Eagle files directly and post process - I tried making Gerbers but ballsed it up and they only charge a couple of bucks to fix up.

Happy to give them to anyone who wants, as long as Rod's ok with it, just PM me.

Regards,
Pete
 
andyjevans said:
As a result of listening tests I and Piotr would rank them as follows:

Rod's design
Current source using LM1084
Voltage regs using LM1084 plus CMC at the end
Voltage regs using LM1084 (no CMC)
Current source using LM1084 plus CMC at the end (didn't sound good)



I'll second that (I never built/listened to Rod's design), I didn't like the final CMC. I'd add that I went a bit further. My set-up is currently:

LCLC for my 801As (and adding the current source at the end didn't change things so I removed it).
CLC + current source for the 211's

Things got better when I added the LC filter before the current source under the 211's

Ciao
Gianluca
 
Any circuit boards still available?

I see that this thread is a few months old, but I just discovered it. Is anyone making or using the PCB's mentioned above?

I would like to get some of these as well, if possible. Can anyone post the specific schematic that was used for making the board layout? The original thread discusses several possibilities and I'd like to know which version was chosen in the end.

Thanks to all,
John
 
Pete's board design is available now, in the form of Eagle Files. These can be emailed, without changes to Olimex -

http://olimex.com/pcb/index.html

PM Pete (posted above) or me for the files.

This design has the advantage of being ready NOW and having been fully tested on a number of boards.

I have said once or twice that I'll do a version, and I am getting near to doing it! It will offer a wide range of power handling for large DHTs, and line-level type, too. I plan to include an interlock signal - able to drive a relay - once the filament is warmed up enough for the voltage to come into spec. (The current-drive filament warms up MUCH slower than any other method). You could use this signal to apply B+ through a relay, since you will get on turn-ON delay of 30~60 seconds, depending on the filament design.

If there are folks interested in something like this, please post with your filament voltage & current, and other considerations, like whether the interlock is a good idea for you. would you like blank boards or built boards?

Note that in all cases, a dc supply (filament rated voltage plus 3 to 5V) will be needed to run the supply, and heatsinking (M3 screws attaching to chassis ideal).
 
Yes - an 813 would be a good challenge. I think these tubes have got into the collections of quite a few of us because of their good value and availability - especially since the 845 seems to have availability problems. Rod didn't design for anything above a 300b as far as I know, and I haven't tried the board on an 813. Heatsinks de rigeur.

Pete and I had a load of these boards made. They're quite tiny and work very well. They were pretty cheap too, so having another batch made is a distinct possibility.

Pete is the designer of the board, and if he has free time he may be able to order another batch and distribute them. If I have any free time I'd try and help him.

These are certainly worthwhile and probably as good as anything out there, so I guess we should get into gear and have some made.

Andy.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.